DETECTABILITY IN CONVENTIONAL AND ADAPTIVE SAMPLING

In this paper a simple but very general method is given for estimating a population total with any sampling design when objects in sampled u nits are observed with imperfect detectability-a problem characteristi c of many surveys of natural and human populations. In the most genera l case, the method consists of dividing the value of the variable of i nterest associated with each detected object by the detection probabil ity for that object and then proceeding to use the estimation method t hat would ordinarily be used under the design if there were no detecta bility problems. Examples illustrating the method include simple rando m sampling, conventional unequal probability sampling, and adaptive cl uster sampling.